pandadoc-followup-list

PandaDoc Unsigned-Document Follow-up List

Find what is still out, work out what deserves attention first, and write the nudge. Run it on a schedule or on demand.

You cannot send the reminder. Be clear about that from the start — see the boundary below.

---

Prerequisites

The PandaDoc MCP Server must be connected. Required tools: PandaDoc:documents_list, PandaDoc:documents_details_get. Optional: PandaDoc:documents_search. If they are missing, say so and point at https://mcp.pandadoc.com.

---

The boundary — what this skill cannot do

There is no reminder endpoint in the PandaDoc MCP, and PandaDoc:documents_send only accepts documents in Draft status. A document that is already Sent therefore cannot be re-sent or nudged through the API.

So this skill surfaces what is stale, ranks it, and drafts the text. The user sends it, or PandaDoc's built-in auto-reminders do. Never say or imply that a reminder was sent. Never call PandaDoc:documents_send on a Sent document — it will fail, and trying suggests you misunderstood what you are doing.

---

Workflow

Step 1: Decide what "waiting" means, and say so

Document status is a closed set of 15. The ones that plausibly mean "out and not finished":

• Sent — delivered, not opened
• Viewed — opened, not signed
• Waiting for Approval — stuck in internal approval, not with the counterparty
• External Review — with the other side's reviewers
• Waiting for Payment — signed but unpaid, a different kind of chase

Default to Sent and Viewed — those are unambiguously "waiting on the counterparty to sign". Include the others only if the user's request implies them ("what's stuck anywhere", "including approvals"), and state in your output which statuses you counted. A list whose scope is invisible cannot be trusted.

Never include Draft (never sent), Completed, Paid, Declined, Expired.

Step 2: Collect them

PandaDoc:documents_list takes a single status string, so this is one call per status. PandaDoc:documents_search accepts status[] but requires a non-empty query, which would silently narrow the sweep — so prefer several PandaDoc:documents_list calls with count: 100.